D-score booklet II
Welcome
1
Introduction
1.1
SDG 4.2.1 indicator
1.2
Quick scan of instruments
1.3
Care-giver vs direct assessment
1.4
Individual-programmatic-population settings
2
Data
2.1
Overview of cohorts and instruments
2.2
Cohort descriptions
2.3
Instruments
3
Comparability
3.1
Are instruments connected?
3.2
Bridging instruments by mapping items
3.3
Age profile of item mappings
4
Equate groups
4.1
What is an equate group?
4.2
Concurrent calibration
4.3
Strategy to form and test equate groups
4.4
Parameter estimation with equate groups
4.5
Common latent scale
4.6
Quantifying equate fit
4.6.1
Well fitting equate groups
4.6.2
Equate groups with poor equate fit
4.7
Differential item functioning
4.7.1
Good equate group without DIF
4.7.2
Poor equate groups with DIF for study
5
Modeling equates
5.1
GCDG data: design and description
5.2
Modeling strategies
5.3
Empirical and fitted item response curves
5.4
Active and non-active equate groups
5.5
Splitting and combining equate groups
5.6
Modeling strategies
5.7
Item information
6
Comparing ability
6.1
D-score distribution by study
6.2
Impact of measurement error
6.3
Domain coverage and scores
6.3.1
Domain coverage of the scale
6.3.2
Domain
\(D\)
-scores
7
SDG 4.2.1 Indicator
7.1
Application I: Estimating the SDG 4.2.1 indicator from existing data
7.2
Definition
developmentally on track
7.3
Country-level estimates
7.4
Relation to other estimates
7.4.1
Stunting
8
Who is on-track?
8.1
Application II: What determines who is developmentally on-track?
8.2
Types of explanatory factors
8.3
Relative importance
8.4
Opportunities for intervention
9
Discussion
9.1
Potential and limitations of existing data for SDG 4.2.1
9.2
Options for improving health policy
9.3
Suitability of D-score metric
9.4
Suggestions for better measurement
References
D-score: Tuning instruments to unitiy
5.3
Empirical and fitted item response curves
(ref:model_1339) D-score by age of four models with all 1339 items using 0, 11, 33 and 184 equate groups.